00:03:13 gilbertw has quit () 02:18:37 abbaJ has quit (Ping timeout for abbaJ[user-33qs21m.dsl.mindspring.com]) 02:25:01 abbaJ (jabba@user-33qs21m.dsl.mindspring.com) has joined #openacs 03:28:44 AaronSw (Snak@c930384-a.hlndpk1.il.home.com) has joined #openacs 06:43:08 abbaJ has quit (Ping timeout for abbaJ[user-33qs21m.dsl.mindspring.com]) 06:56:31 abbaJ (jabba@user-v3qs3p0.dsl.mindspring.com) has joined #openacs 07:21:05 abbaJ has quit (Client Exiting) 07:21:14 abbaJ (jabba@user-v3qs3qv.dsl.mindspring.com) has joined #openacs 07:38:01 djg (dirk@pD9001D79.dip.t-dialin.net) has joined #openacs 09:12:57 ola (ola@as3-1-3.mal.s.bonet.se) has joined #openacs 09:13:02 rzolf (rolf@adsl-64-167-240-171.dsl.snfc21.pacbell.net) has joined #openacs 09:15:28 anyone alive 09:19:13 ya 09:19:34 if i have a table with columns key, language, message 09:19:52 and i want to display a result: 09:20:07 key, english_translation, japanese_translation, message 09:20:24 that leaves english_translation or japanese_translation null 09:20:29 if it isn't there 09:20:34 how do i do that. 09:20:40 i have a self join now 09:20:43 that looks like this: 09:21:22 eek that violates normal form 2 (or 3) 09:21:47 this is always ugly if you don't have decode 09:21:56 select m.key, m.message as japanese_translation, m1.message as english_translation from lang_messages m left join lang_messages m1 on (m.key = m1.key and m.lang='ja' and m1.lang='en' group by m.key, jt, et; 09:22:02 hm 09:22:13 put a right paren before the group by. 09:22:27 that is working ok, but giving me duplicates 09:22:31 u want either a japanese or an english message? 09:22:51 i want a table that lists the translation for each key 09:23:03 and leaves a blank if one of the translations are missing 09:23:16 why not like this: message_id, language_id, message 09:23:32 and then outer join between message and translation table 09:23:36 with nvl(message, "") 09:23:53 the table already exists 09:23:59 with data in it 09:24:14 it is ACS lang, btw. ;-) 09:24:16 select nvl(message,"") from translations t, messages m where m.message_id=t.message_id and t.language_id=the_one_you_want; 09:24:36 yeah i was lucky to use english-language systems only ;) 09:26:07 the thing is i want both languages 09:26:13 concatenated into one row. 09:26:28 hmm, i still dont understand 09:26:37 that looks like 09:26:46 the table contains data that looks like this" 09:27:00 in which directory is this table? 09:27:04 key, lang, message 09:27:06 i mean...where is the create table 09:27:14 hold 09:27:24 ok and key and lang are concatenated primary key columns? 09:27:40 it might not be ported yet 09:27:55 the data looks like 09:28:13 admin_header, en, "administration" 09:28:25 admin_header, jp, "japanese word for administration" 09:28:38 i want to do a select that produces: 09:28:42 ah now i get it 09:28:55 admin_header, "administration", "japanese word for administration" 09:29:27 it should be possible with an outer self join 09:29:37 but i am doing something wrong 09:30:43 select m1.key, m1.message, m2.message 09:30:43 from messages m1, messages m2 09:30:43 where m1.key='admin_header' and m1.language_code='engrish' 09:30:43 and m2.key='admin_header' and m2.language_code='japanese' 09:30:54 or m2.key=m1.key if you insist on a join 09:31:07 ah 09:31:11 hold on 09:33:02 ok, so now how do i generate a null in a column for those keys 09:33:08 that are missing one of the translations 09:33:18 and group it together so i don't have dupes 09:33:58 w8 a second 09:34:10 is there nvl in postgres? 09:34:39 yes there os 09:34:39 there is case 09:35:15 what is case? 09:35:20 decode 09:35:26 ok 09:35:33 why would you get dupes? 09:35:40 i dont know. 09:35:54 oops 09:35:55 where rownum=1 helps against dupes in oracle hehe 09:36:08 rzolf has quit (rzolf has no reason) 09:36:14 rzolf (rolf@adsl-64-167-240-171.dsl.snfc21.pacbell.net) has joined #openacs 09:36:21 oops 09:36:32 don't use emacs commands in BitchX, i guess. 09:37:07 me? 09:37:12 no me ;-) 09:37:16 ok 09:37:37 i am doing this: 09:37:45 can u point me to the create table statement? 09:37:49 i need to try this 09:38:04 but i think you're taking a very complicated approach ;) 09:38:13 it should just work. 09:38:40 yes 09:38:45 try my query 09:39:00 what is the problem about this query? 09:39:02 your query only returns where key=key 09:39:10 it doesnt fill in nulls. 09:39:20 create table lang_messages ( 09:39:25 key varchar(800), 09:39:31 lang varchar(20) not null, 09:39:34 message text, 09:39:55 hmm so if cannot find a row you still want one row with nulls? 09:40:08 if 09:40:21 admin_header has a japanese translation 09:40:26 but not an english translation 09:40:46 ah 09:40:48 i want to get: admin_header, "japanese translation", "" 09:41:09 currently 09:41:11 this query" 09:41:13 select m.key, m.message as j_translation, m1.message as e_translation f\ 09:41:14 rom lang_messages m left join lang_messages m1 on m.key=m1.key and m.lang = 'ja' \ 09:41:15 and m1.key = 'en' group by m.key, j_translation, e_translation; 09:41:22 does almost what i want. 09:42:22 wait that is wrong 09:42:25 the wrong query 09:42:50 is one of the languages completely translated? 09:42:57 no. 09:42:57 like...can i use outer joins at all? 09:43:42 maybe i can just use decode and a group by? 09:43:54 no, i think i would do this in the application 09:44:05 hehe 09:44:11 that sucks, though. 09:44:14 gimme the english and japanese translataion 09:44:15 if not there gimme the japanese translation 09:44:15 the TCL cop-out 09:44:20 yeah, it sucks 09:44:30 i have to think about it 09:45:08 i think decode and group by will work. 09:45:15 maybe, but it lokks very kludgy 09:45:23 and that for a straight-forward thing 09:45:44 it should be an easy self join 09:46:00 unless it doesnt work, because there are missing values in each column 09:46:03 instead of just one 09:48:57 but it sounds like you need control-flow 09:49:06 sou tcl aint that much of a cop-out 09:53:38 i dont uderstand why i am getting a duplicate. 09:54:29 select distinct for the lame 09:54:37 but that's yet another kludge 09:54:38 that's no good. 09:55:24 for what is the group by? you dont do agregates 09:55:46 true 09:55:59 where is the duplicate coming from. 09:56:15 and why is someone on my machine sending me "talk" requests. 09:56:19 that's so 1980s... 09:56:30 try each query withut the join 09:57:04 that's very strange that you get a duplicated row 09:57:07 as a result 09:58:35 so if i leave off the join 09:58:37 and just do 09:59:04 well just encapsulate for now 09:59:06 select m.key, m.message, m1.message from lang_messages m, lang_messages m1 where m.lang='ja' and m1.lang='en' 09:59:15 and think about it for a few days 09:59:32 i get the proper number of rows returned 10:00:37 hmm 10:00:43 but 10:00:46 i dont understand it either 10:06:39 l8r 10:06:42 lunch time 10:08:11 rzolf has quit (rzolf has no reason) 10:27:32 ola has quit (ircII EPIC4-1.0 -- Are we there yet?) 11:09:30 rzolf (rolf@adsl-64-167-240-171.dsl.snfc21.pacbell.net) has joined #openacs 11:19:41 rzolf has quit (rzolf has no reason) 12:09:50 rzolf (rolf@adsl-64-167-240-171.dsl.snfc21.pacbell.net) has joined #openacs 12:22:59 re 12:23:05 solved yer problem? 12:37:35 markd2 (markd2@r-41.127.alltel.net) has joined #openacs 12:37:48 jmarsden (jonathan@host213-122-49-74.btinternet.com) has joined #openacs 13:09:30 markdfoo (markd2@r-41.127.alltel.net) has joined #openacs 13:09:52 * markdfoo does not like his isp 13:10:37 markd2 has quit (Ping timeout for markd2[r-41.127.alltel.net]) 13:11:13 markdfoo is now known as markd2 13:31:23 jmarsden has quit (Ping timeout for jmarsden[host213-122-49-74.btinternet.com]) 13:56:27 markd2 has quit (Bork) 14:37:46 jmarsden (jonathan@host213-122-71-102.btinternet.com) has joined #openacs 14:39:37 til (yann@62.116.25.202) has joined #openacs 14:43:19 morning all 14:50:18 brb 14:52:50 markd2 (markd2@r-41.51.alltel.net) has joined #openacs 15:18:10 Is jmarsden around? 15:31:44 jmarsden has quit (Ping timeout for jmarsden[host213-122-71-102.btinternet.com]) 15:37:04 I guess not anymore :) 15:41:59 jmarsden (jonathan@host213-122-40-13.btinternet.com) has joined #openacs 15:42:31 rbm 15:43:14 rbm: I'm back at the computer... 15:43:24 hey jonathan 15:43:42 I just noticed your RPMs were not linked from the software page! What a shame on me! 15:44:07 So I just linked them there 15:44:15 Someone else said something about that in a bboard post... cool. 15:44:45 I'm preparing a list of contributors to the project. we need to recognize people better.p 15:44:52 I'm looking at redoing the aolserver one using the newer aolserver 3.3+ad13 that aD has at their site, and also making it so libxml2 is not needed for OpenACS 3.x 15:45:23 Yes... I said that to Ben in my response to his request for comments on the 'leadershi' stuff recently. 15:45:56 He asked me to prepare this list at least 2 weeks ago. 15:46:01 Not that I need recognition but some of the porters for example I think it woudl be really good to recognize "officially". 15:46:18 I'm glad Musea is taking over the site. It'll be really good for everyone. 15:46:37 Ah, OK... yes, should give you mroe time for docs and improve the site too. 15:46:51 jmarsden: I agree with you. Recognition is crucial, especially in free software projects 15:48:07 Maybe we can at elast start a bboard trhead when someone ports a module, congratulating them and encouraging the rest of the cmmunity to test it? 15:49:35 bbaquiran (bpb@202.8.232.55) has joined #openacs 15:50:24 That'd be good. But it depends on the porters to start the thread, since we're not monitoring who finishes what. Well, Don is. 15:50:52 RIght... maybe it's even betetr if Don starts those trheads, as the project leader! 15:54:00 zzzirk_slc (lzirkel@166.70.157.42) has joined #openacs 15:54:07 true 15:54:17 moo rbm 15:54:18 Send Don an e-mail about that. He'll like your suggestion. 15:54:26 Ok, sure. 15:54:27 moo zzzirk 15:54:35 whassup? 15:54:38 zzzirk: hey, did you get the CR install issue licked? 15:54:43 Does anyone know if opn support memos? 15:55:44 Nope, not quite. I just did a CVS update and tried a new installation from scratch and I'm down to 8 of those *ABORT STATE* statements. There were a TON more yesterday. 15:56:24 Sounds like time to post the relevant chunks from the log to the bboard and see who else can help. 15:56:28 steveW (swoodcock@194.72.81.226) has joined #openacs 15:56:38 I didn't know if maybe other people were doing upgrades from OpenACS 3.2 and I'm doing a fresh install. 15:57:15 You don't upgrade the datamodel anyway, you run the thing as a separate PG user and so forth, so I dobt that's it. 15:57:16 Yes. OPN supports memos. That's really cool. 15:57:58 help note 15:58:31 I didn't think so either, that was just the ONLY thing I could think of why other people weren't having problems. 15:59:08 Use the power of the community and post the errors asking for help... *someone* will know how to solve it :-) 15:59:50 I was actually planning to do some investigation and post a message last night, but the time never materialized in which to do anything. I'm hoping to get to it today or tonight. 16:03:59 * rbm prepares to do hmwk for his mega-lame class 16:04:18 BTW, the "tramp" Emacs module is awesome 16:04:22 Perl class? ;-) 16:04:29 Worse. PHP. 16:04:36 OOoooooh. :-) 16:04:48 But that's not the worse part. The professor is the worst part. 16:05:18 I don't know how can people develop big things in PHP. It must be an exercise of oriental patience. 16:06:41 I tried using persistent connections under PHP, and they are _really_ persistent. They never get re-used. With 2 users pounding on my machine I have over 40 connections open after 2 minutes. 16:07:09 sounds like a driver bug 16:07:51 Perhaps. It was the PG driver. I'm not that familiar with the PHP internals, but if it's a bug, it's been there for a loooong time, since everybody counsels NOT to use persistent connections. 16:09:16 I also tried using the PEAR DB abstraction layer only to find out it took 5 minutes to grab a connection. I had to come up with my own db abstraction layer (that coincidently ended up resembling AOLserver's API :-)) 16:10:26 bbaquiran has quit (Ping timeout for bbaquiran[202.8.232.55]) 16:18:43 zzzirk_slc: are you still there? 16:19:00 I am. 16:19:11 I've been thinking about your installation problem 16:19:24 Are you near your machine? 16:19:30 Oh, what are you thoughts? 16:19:35 Physically close? No. 16:20:31 I had the same problem - acs-content-repository failing the install, plus one other package (can't remember which) 16:20:53 Anyway, trying to fix something else I cut down the number of records in tree_encodings 16:20:59 And the next time the install worked 16:21:57 Didn't know whether I was seeing things, but just did the same thing here 16:23:59 I reduced tree_encoding to 0-9, A-Z 16:24:28 I have no idea why this affects the CR install... 16:25:39 Where do you set tree_encoding? Is that a database setting? 16:26:07 Its in acs-kernel/sql/postgresql/postgresql.c 16:26:35 Its used to generate the tree_sortkey field on the "hierachical" tables 16:26:50 Do you mean postgresql.sql? 16:26:59 Ack, yes 16:27:16 Okay and you reduced it to 0-9, A-Z right? 16:27:33 No other punctuation or lower case letters? 16:27:36 Yup. Just now (at work) I cut it down to 0-9 16:27:58 The idea is you want "select * from tree_encodings order by code" to come out in "deci" order 16:28:27 Right now I've code "deci" 0-9 and "code" 0-9 16:28:47 At home I had "deci" 0-35 and "code" 0-9A-Z 16:29:26 So I should renumber the alphabet characters as well? 16:29:34 As far as the number on the left? 16:29:44 I did 16:30:15 Okay, I've made that change. I'm gonna try it, hang on. 16:30:16 You want to change the hardcoded number in tree_default_encoding_base() as well (just underneath the set of tree_encoding rows) 16:30:29 ...to the number of records you've got now 16:31:46 36 in other words. 16:33:41 Ok, the install is running now. 16:34:01 Moment of truth. 16:34:42 Everything went through just fine. It installed. 16:34:57 Thanks stevew! 16:35:51 This is just a sticking plaster though... 16:36:10 And I think it buggers up tree_sortkey in a slightly different way 16:36:37 That is *weird*! Let Don and whoever ported acs-kernel know about this one 16:37:08 Can you do a "select object_type, tree_sortkey from acs_object_types order by tree_sortkey"? 16:38:01 For acs_mail_link I get /00/100 16:38:10 Should be /00/10 I think 16:41:10 I'll mail Don & co. 16:42:45 I got /00/0a 16:47:42 Oh, I used 0-9. That looks right for you 16:47:51 I wonder what you get after /00/0z? 16:48:27 try this: select tree_next_key('/0z'); 16:49:14 openacs-4=# select tree_next_key('/0z'); 16:49:14 tree_next_key 16:49:14 --------------- 16:49:14 16:49:14 (1 row) 16:49:32 That's what I got back. 16:49:50 how about for /0y ? 16:50:05 Okay, so maybe I should have uppercased my A the first time. It was /00/0A not /00/0a 16:50:20 ah, ok 16:50:24 Same thing. It must want upper case. 16:50:31 yes 16:51:44 Can you try it with /0Z? 16:53:29 I did try it with /0Z and got /100 16:53:54 openacs-4=# select tree_next_key('/0Z'); 16:53:54 tree_next_key 16:53:54 --------------- 16:53:54 100 16:53:54 (1 row) 16:54:19 Ok, thanks. I'll have a look into it tonight... 16:54:33 I wonder what's different with mine and your installations... 16:54:43 Is that something you're working on? 16:54:47 I don't know. Did you get something different? 16:55:29 I mean why did we have a problem, but noone else? 16:58:27 I don't know. I've been wondering the same thing. Did you install PG from RPMs or source? 16:59:12 RPMs (7.1.2PGDG) 16:59:16 Have you been able to install any other packages? I've tried bboard and ticket tracker and both of them seem to install and then bomb during use. 16:59:31 I installed from the same. 16:59:59 Nope. Are they finished? 17:00:29 I don't know, I just thought I'd give 'em a try and see what happened. I just didn't know if you'd tried anything else at all. 17:02:07 Nope, I'm trying to get acs-subsite finished... 17:03:28 Is that what you're working on then? 17:03:28 shagster has quit (Read error to shagster[web1.alal.com]: Connection reset by peer) 17:04:21 Yes. 17:04:55 Where I work we are deploying an ACS Classic site before long (hopefully) and then we'll be doing a couple of internal projects that we're hoping to use OpenACS for. We've been wanting to help with the porting effort, but this site redesign we're in the middle of has prevented that. 17:06:09 It's hard to find the time to work on it these days 17:07:10 Doesn't that always seem to be the case. 17:07:49 Speaking of which, I've got to go and get the dinner on! 17:09:02 bbl 17:09:05 steveW has quit (Read error to steveW[194.72.81.226]: EOF from client) 17:11:00 * rbm returns after a trip to the world of Linux-based PDAs 17:15:12 zzzirk_slc has quit ([x]chat) 17:17:42 GEM has quit (Ping timeout for GEM[MULTICS.MIT.EDU]) 17:18:20 GEM (mcmullan@MULTICS.MIT.EDU) has joined #openacs 17:34:19 jmarsden has quit (Ping timeout for jmarsden[host213-122-40-13.btinternet.com]) 18:22:43 djg has quit (Ping timeout for djg[pD9001D79.dip.t-dialin.net]) 18:22:50 djg (dirk@pD9001D79.dip.t-dialin.net) has joined #openacs 18:33:34 and then outer join between message and translation table/quit 18:33:40 djg has quit (Read error to djg[pD9001D79.dip.t-dialin.net]: EOF from client) 18:44:26 markd2 has quit (Ping timeout for markd2[r-41.51.alltel.net]) 18:53:26 til has quit ([x]chat) 18:57:12 markd2 (markd2@r-41.51.alltel.net) has joined #openacs 19:03:19 shagster (mkovach@web1.alal.com) has joined #openacs 19:35:58 shagster has quit (Ping timeout for shagster[web1.alal.com]) 19:37:32 shagster (mkovach@web1.alal.com) has joined #openacs 19:42:13 shagster has quit (Ping timeout for shagster[web1.alal.com]) 19:45:31 jmarsden (jonathan@host213-122-119-220.btinternet.com) has joined #openacs 20:02:31 shagster (mkovach@web1.alal.com) has joined #openacs 20:41:25 ola (ola@as3-1-3.mal.s.bonet.se) has joined #openacs 21:15:23 shagster has quit (Read error to shagster[web1.alal.com]: Connection reset by peer) 21:19:58 jmarsden has quit (Leaving) 21:49:56 GEM has quit (Ping timeout for GEM[MULTICS.MIT.EDU]) 21:51:31 GEM (mcmullan@MULTICS.MIT.EDU) has joined #openacs 22:03:27 moo?? 22:04:31 * markd2 meditates. oom 22:10:34 markd2 has quit (dinner) 22:43:04 So are people getting hit by the IIS work already? 22:43:07 I got 21 hits today. 22:43:15 s/work/worm/ 22:47:37 Oh, is that what that is? 22:54:04 Got hit 70 times today 22:56:35 rbm has quit (Ping timeout for rbm[slxwy.dorms.usu.edu]) 23:01:42 ola has quit (ircII EPIC4-1.0 -- Are we there yet?) 23:03:57 rbm (roberto@slxwy.dorms.usu.edu) has joined #openacs 23:04:11 AaronSw: Yeah. It's an IIS worm. And it'll be _really_ nasty 23:04:37 How so? 23:04:58 Starting tomorrow, an estimated 200,000 infected hosts will be sending 410 Mb of data each to whitehouse.gov every 4.5 hours 23:05:12 eek 23:05:24 Well, that'll be fun. 23:05:26 So the 'net will be crawling tomorrow. 23:05:32 (presumably) 23:05:48 I'll probably have to turn off the email alerts... 23:05:54 heh 23:06:02 Time to go to class 23:06:05 c'ya 23:06:08 bb in 3 hours 23:06:16 whish me luck in my test :) 23:06:23 best of luck 23:06:25 :-) 23:06:28 Tks :) 23:13:19 abbaJ has quit (Ping timeout for abbaJ[user-v3qs3qv.dsl.mindspring.com]) 23:13:54 shagster (mkovach@web1.alal.com) has joined #openacs 23:45:27 GEM has quit (Ping timeout for GEM[MULTICS.MIT.EDU]) 23:53:43 derfy (derfy@fcy-5.soho.enteract.com) has joined #openacs